Discovering Adaptable Symbolic Algorithms From Scratch

# Discovering Adaptable Symbolic Algorithms from Scratch

**arXiv ID:** 2307.16890
**Authors:** Stephen Kelly, Daniel S. Park, Xingyou Song, Mitchell McIntire, Pranav Nashikkar, Ritam Guha, Wolfgang Banzhaf, Kalyanmoy Deb, Vishnu Naresh Boddeti, Jie Tan, Esteban Real
**Published:** 2023-07-31T17:57:48Z
**Abstract:**
Autonomous robots deployed in the real world will need control policies that rapidly adapt to environmental changes. To this end, we propose AutoRobotics-Zero (ARZ), a method based on AutoML-Zero that discovers zero-shot adaptable policies from scratch. In contrast to neural network adaptation policies, where only model parameters are optimized, ARZ can build control algorithms with the full expressive power of a linear register machine. We evolve modular policies that tune their model parameters and alter their inference algorithm on-the-fly to adapt to sudden environmental changes. We demonstrate our method on a realistic simulated quadruped robot, for which we evolve safe control policies that avoid falling when individual limbs suddenly break. This is a challenging task in which two popular neural network baselines fail. Finally, we conduct a detailed analysis of our method on a novel and challenging non-stationary control task dubbed Cataclysmic Cartpole. Results confirm our findings that ARZ is significantly more robust to sudden environmental changes and can build simple, interpretable control policies.

## Skill Description

This skill is generated from the arXiv paper: Discovering Adaptable Symbolic Algorithms from Scratch (2307.16890).

## How to Use

[To be filled in by the user or by future automation]

## References

- [arXiv:2307.16890](http://arxiv.org/abs/2307.16890v2)